Please correct my work. The textbook answer which is expressed exactly like this $1/s(1+e^{-s})$ does not match my own.

enter image description here

Find the Laplace Transform for one period of the perpetual periodic function shown in graph.

First express function using the unit step:

$$f(t)= u(t)-u(t-1)$$

The Laplace Transform is easily found:

$$f(t)= \frac{1}{s}-\frac{e^{-s}}{s}= \frac{1-e^{-s}}{s}$$

Where did I go wrong?


2 Answers 2


Here is a very simple method to solve this problem. Note that $f(t)-f(t-2)$ is just a single period of $f(t)$ in the range $[0,2]$, i.e.


the Laplace transform of which you've already found:


But you also have


where $F(s)$ is the desired Laplace transform of the periodic function $f(t)$. Equating (2) and (3) gives





Your $f(t)$ is correct for the imagine provided if you only take one period. $$ f(t) = \mathcal{U}(t)-\mathcal{U}(t-1) $$ Now take the Laplace transform of $$ \int_0^{\infty}f(t)e^{-st}dt = \int_0^{1}e^{-st}dt $$

If you want to find the Laplace transform of the periodic unit step function, then first we will write $f(t)$ as a series. $$ f(t) = \sum_{n=0}^{\infty}\bigl[\mathcal{U}(t-2n)-\mathcal{U}[t-(2n+1)]\bigr] $$ Let's write out a few terms. \begin{align} f(t) &= \mathcal{U}(t)-\mathcal{U}(t-1)+\mathcal{U}(t-2)-\mathcal{U}(t-3)+\mathcal{U}(t-4)-\cdots\\ & = \mathcal{U}(t) + \sum_{n=1}^{\infty}(-1)^n\mathcal{U}(t-n)\\ \mathcal{L}\{f(t)\} &= \int_0^{\infty}e^{-st}dt + \int_0^{\infty}\sum_{n=1}^{\infty}(-1)^n\mathcal{U}(t-n)e^{-st}dt \end{align} We can interchange the $\sum(-1)^n$ with integral since $\mathcal{U}(t-n)e^{-st}\geq 0$ for $t \geq 0$. \begin{align} &= \frac{1}{s} + \sum_{n=1}^{\infty}(-1)^n\int_n^{\infty}e^{-st}dt\\ &= \frac{1}{s} + \sum_{n=1}^{\infty}(-1)^n\frac{e^{-sn}}{s}\\ &= \frac{1}{s}\Bigl[1 - e^{-s}\sum_{n=0}^{\infty}(-1)^ne^{-sn}\Bigr]\\ &= \frac{1}{s}\Bigl[1 - \frac{e^{-s}}{1+e^{-s}}\Bigr] \end{align}

  • $\begingroup$ I assumed I only needed one period to solve it correctly but if that is not correct how should I go about solving for the entire function (to infinity)? At best my textbook does a very sub par explanation of the unit step function. $\endgroup$ Feb 3, 2015 at 3:00
  • $\begingroup$ Thank you again. I had thought I may have to use a summation series but I was hoping for something simpler like an integral. $\endgroup$ Feb 3, 2015 at 3:06
  • $\begingroup$ sorry please disregard this comment $\endgroup$ Feb 3, 2015 at 3:09
  • $\begingroup$ I got the idea to use just one period (cycle) from this formula: $$ℒ\{f(t)\}=\frac{ℒ\{f_​1​​(t)\}}{1−e​^{−sp}}$$ where p is the period and $f_1$ is the function for one period. I tried it but I could not get the same answer as in text. I found it at intmath.com/laplace-transformation/… $\endgroup$ Feb 3, 2015 at 3:14

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .